电子学报 ›› 2019, Vol. 47 ›› Issue (12): 2582-2589.DOI: 10.3969/j.issn.0372-2112.2019.12.018

• 学术论文 • 上一篇    下一篇

一种基于动态显著性的人眼注视点移动计算模型

李鹏1,2, 王延江1   

  1. 1. 中国石油大学(华东)信息与控制工程学院, 山东青岛 266580;
    2. 中国石油大学(华东)计算机与通信工程学院, 山东青岛 266580
  • 收稿日期:2018-08-13 修回日期:2019-06-11 出版日期:2019-12-25
    • 通讯作者:
    • 王延江
    • 作者简介:
    • 李鹏 男,1975年出生,山东昌邑人,中国石油大学(华东)博士研究生,讲师.研究方向为计算机视觉、模式识别.E-mail:lipieu2002@163.com
    • 基金资助:
    • 国家自然科学基金资助项目 (No.61271407,No.61671480); 山东省自然科学基金面上项目 (No.ZR2017MF069,No.ZR2018MF017)

A Computational Model of Fixation Movement Based on Dynamic Saliency

LI Peng1,2, WANG Yan-jiang1   

  1. 1. College of Information and Control Engineering, China University of Petroleum(East China), Qingdao, Shandong 266580, China;
    2. College of Computer and Communication Engineering, China University of Petroleum(East China), Qingdao, Shandong 266580, China
  • Received:2018-08-13 Revised:2019-06-11 Online:2019-12-25 Published:2019-12-25
    • Corresponding author:
    • WANG Yan-jiang
    • Supported by:
    • National Natural Science Foundation of China (No.61271407, No.61671480); General Program of Natural Science Foundation of Shandong Province,  China (No.ZR2017MF069, No.ZR2018MF017)

摘要: 人类对各种复杂视觉场景的感知通过注视点的不断移动来实现.但在计算机视觉中,如何准确模拟人眼注视点的移动而实现"看向哪儿"这一生物行为是一个难题.为此,本文基于动态显著性提出一种自由观察情形下的注视点移动计算模型.模型包括了全局跃迁和局部转移行为,前者是通过将眼动趋势、向心性和返回抑制性等规则进行融合来预测下一个注视点从而实现相对远距的注视点迁移;后者则是通过查找局部最大平均显著性点并将其作为转移的目的位置,二者通过预设准则进行切换.实验表明,所提模型能够更有效地模拟人眼注视点分布,具有良好的适用性.

关键词: 视觉显著性, 注视点, 视线, 跃迁, 转移

Abstract: The human eyes observe and perceive different external scenes by the continuous movement of fixations. It is of great difficulty to predict and model "where to look" of human eyes in computer vision research. To address this problem, a computational model including two parallel processes: global transition and local movement, is proposed based on visual dynamic saliency. The former predicts the long distance migrations of saccades in the scene by combining the eye movement bias,visual centrality and inhibition of return; the latter is used to determine the next destination by searching for the local largest average salient point. And these two processes can be switched through a certain judgment rule. Experimental results show that the proposed model can more effectively simulate the human eye gaze point distribution, resulting in better performance.

Key words: visual saliency, fixation point, gaze, transition, shift

中图分类号: